A study on optimal selection of reference accelerometers for active road noise control

Active noise control (ANC) is gradually expanding its application in the automotive industry. Passive noise control (PNC) methods are not advantageous for lightweighting as they increase the weight of the vehicle, and ANC is more effective to control low frequency range noise as compared to PNC methods. In terms of ANC adaptation, the performance of ANC systems is directly affected by the selection of the reference signal. This study suggests a method that determines candidates for the locations of the reference signal sensors by using the correlation between the input acceleration signal and output sound pressure level signal, which is a partial coherence function. This process evaluates the correlation of each acceleration signal with the interior noise and eliminates redundant sensor locations by maximizing the Fisher information matrix. This method enables the reduction of the total number of candidate locations, which in turn leads to a decrease in the total time cost while obtaining locally optimal results.